(3 1. X. 94) The objectives of this study were to gain insights into the structure-lipophilicity relationships of peptides and to propose an improved model for estimating their lipophilicity. First, existing databases were extended to obtain the distribution coefficients of a total of 208 free or protected peptides (di-to pentapeptides). The polarity parameters ( A j of 23 free amino acids and 19 protected amino acids (AcNH-CHR-CONH,) and of their side chains were Calculated from experimental distribution coefficients and computed molecular volumes. An analysis of the polarity parameters revealed that the hydrophobicity of the amino-acid side chains is largely reduced due to the polar field of the backbone. The polarity parameters of the peptides were then obtained in a similar manner and shown to be highly correlated with the sum of the polarity parameters of their side chains, i.e., the lipophilicity of peptides can be calculated from their molecular volume and the sum of their side-chain polarities using the regression established for each individual series of peptides (Fig. I). This last restriction is essential since the polarity and lipophilic increment of a NH-C*H-CO unit were shown to decrease with increasing length of backbone.
Introduction.-Endogenous peptides such as many hormones and neurotransmitters were found to modulate a wide variety of biological functions [ 11 121. The potency and specificity of these endogenous compounds make it clear why the design of synthetic peptides and of peptidomimetics is already one of the major issues in drug research. Indeed, these approaches led to the discovery of new lead compounds and drugs such as potent peptide receptor antagonists and enzyme inhibitors.Peptides and their constituent amino acids show a wide range of physicochemical and structural properties. Amino-acid side chains contain polar, non-polar, charged, or uncharged groups and differ considerably in size and flexibility. Characterizing the structural and physicochemical properties of amino acids and peptides is an important condition in molecular biology to unravel the properties of proteins, and in molecular pharmacology to rationalize and predict the biological properties of peptide drugs. What is at stake here is a better understanding of structure-property-activity relationships of peptides. In this context, lipophilicity parametrization of amino acids and peptides is of major concern. Indeed, lipophilicity is a physicochemical property of particular significance in drug design, because it encodes a wealth of information on a solute's structure and the intermolecular interactions it elicits [3-51.Lipophilicity can be expressed by the logarithm of the partition coefficient (i.e. log P, which refers to a well-defined electrical state of the solute, for example the neutral or